Long-term treatment with anti-VEGF does not induce cell aging in primary retinal pigment epithelium

Anti-Vascular Endothelial Growth Factor (VEGF) therapy is given repeatedly for an extended period of time to patients when treated for age-related macular degeneration. While short-term effects of anti-VEGF agents on retinal pigment epithelial (RPE) cells have been investigated, the effects of long-term and repeated treatment on these cells are scarce. In this study, we have investigated the effects of anti-VEGF treatment after long-term, repeated treatment on cell aging and morphology. The experiments were conducted in primary porcine RPE cells passage one and two. Cells were treated with 125 μg/ml bevacizumab, ranibizumab, aflibercept or rituximab once a week for 1 day, 4 days, 7 days, 4 weeks and 12 weeks. Cell survival was evaluated with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide tetrazolium (MTT) and trypan blue exclusion assay. Activity of β-galactosidase was assessed in a commercially available assay. Influence of these compounds was investigated on the expression of cathepsin D and amyloid β, and the expression and phosphorylation of mechanistic target of rapamycin (mTOR), all proteins involved in senescence and aging, in Western blot. The secretion of Pigment Epithelium Derived Factor (PEDF) and Transforming Growth Factor (TGF)-β was investigated in Enzyme-linked Immunosorbent Assay (ELISA). The cellular morphology was investigated with electron microscopy, investigating the number and area of mitochondria and autophagosomes. Statistical analysis was conducted using a mixed linear model. Weekly treatment up to 12 weeks displayed no toxic effects on RPE cells in any of the substances tested. Ranibizumab showed a significant increase in β-galactosidase signal on day 4 (p < 0.05) and 7 (p < 0.05) after treatment. In long-term, however, ranibizumab displayed no significant difference to untreated cells. Bevacizumab displayed a significant reduction of the β-galactosidase signal after 12 weeks (p < 0.05). Aflibercept significantly decreased β-galactosidase after 1 day (p < 0.01) and 12 weeks (p < 0.05). Rituximab and bevacizumab also decreased β-galactosidase signal after 12 weeks (p < 0.05). The expression of mTOR, phospho-mTOR, amyloid β and cathepsin D was not significantly altered by any of the compounds tested. RPE cells secreted considerate amounts of TGF-β. Bevacizumab treated cells showed significantly lower TGF-β secretion than ranibizumab and rituximab (p < 0.05). In contrast, only small amounts of PEDF were secreted which were not altered by any substance tested. Ultrastructural analysis showed no alterations in mitochondria after long-term treatment with either substance. Autophagosomes were not reduced by long-term anti-VEGF treatment compared to control. However, the area of autophagosomes in bevacizumab and aflibercept treated cells was significantly less compared to both ranibizumab and rituximab treated cells (all p < 0.05). Taken together, weekly treatment with VEGF-antagonists up to 3 months does not induce premature aging in primary RPE cells in any tested compound. A significant difference can be found between bevacizumab and aflibercept on the one hand, and ranibizumab (and rituximab) on the other hand, with more autophagosomal area in ranibizumab and (rituximab). Taken together, our data provide indications for long-term safety of anti-VEGF compounds. Further research is warranted.